[netCDFDecoders #DRY-582075]: nam 211 cdl

Daryl,

Here's the basic idea on a setup. The LDM is used to store the raw grib files
because the netCDF decoded ones are about 10-50 times larger and most folks
only want a selected number of parameters out of the file. The raw files are
indexed using either Grib1Indexer or Grib2Indexer routines out of the library
via a crontab routine every 10 minutes to keep index current with the files.
By reading the index, one can use Grib1GetData and Grib2GetData routines to
extract the desired data from the raw file. That's how the THREDDS Data 
Server(TDS) works or the ToolsUI program. If you are using the IDV or the
ToolUI programs, then you can connect to the TDS or local data directory to get 
the data. There are other routines to convert the complete model to netCDF but 
we don't suggest that approach since some of the raw model files are huge, ie 
.5 degree Global GFS model raw file is 2.5 gigs are the decoded one in much 
larger. 

So the main question to you, is what s/w are you planning to use to display the 
data?   It's not hard to setup a local TDS server or use the ToolsUI programs. 
I must mention that the ToolsUI program is basically a debugging
tool so the displays are some what crude. There is a TDS tutorial located at:

http://www.unidata.ucar.edu/projects/THREDDS/tech/tutorial/schedule.html

with a detail section on TDS/IDD configuration and ToolsUI download at:
http://www.unidata.ucar.edu/software/netcdf-java/v2.2/webstart-dev/index.html

if one uses the url in the THREDDS tag of ToolsUI you can see the catalog
http://motherlode.ucar.edu:8080/thredds/catalog.xml

or in a brower use:

http://motherlode.ucar.edu:8080/thredds/catalog.html

I went into detail about this since I didn't know you understanding about
this approach, hopefully I wasn't reiterating info to you.

Robb...

Ticket Details
===================
Ticket ID: DRY-582075
Department: Support netCDF Decoders
Priority: Normal
Status: Open